US4563737A - Virtual storage management - Google Patents

Virtual storage management Download PDF

Info

Publication number
US4563737A
US4563737A US06/448,274 US44827482A US4563737A US 4563737 A US4563737 A US 4563737A US 44827482 A US44827482 A US 44827482A US 4563737 A US4563737 A US 4563737A
Authority
US
United States
Prior art keywords
address
area
virtual
program
task
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Expired - Lifetime
Application number
US06/448,274
Other languages
English (en)
Inventor
Tomoaki Nakamura
Keiichi Nakane
Hiroaki Nakanishi
Koji Hirai
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hitachi Ltd
Original Assignee
Hitachi Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hitachi Ltd filed Critical Hitachi Ltd
Assigned to HITACHI, LTD., 5-1, MARUNOUCHI 1-CHOME, CHIYODA-KU, TOKYO, JAPAN, A CORP OF JAPAN reassignment HITACHI, LTD., 5-1, MARUNOUCHI 1-CHOME, CHIYODA-KU, TOKYO, JAPAN, A CORP OF JAPAN ASSIGNMENT OF ASSIGNORS INTEREST. Assignors: HIRAI, KOJI, NAKAMURA, TOMOAKI, NAKANE, KEIICHI, NAKANISHI, HIROAKI
Application granted granted Critical
Publication of US4563737A publication Critical patent/US4563737A/en
Anticipated expiration legal-status Critical
Expired - Lifetime legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F12/00—Accessing, addressing or allocating within memory systems or architectures
    • G06F12/02—Addressing or allocation; Relocation
    • G06F12/08—Addressing or allocation; Relocation in hierarchically structured memory systems, e.g. virtual memory systems
    • G06F12/10—Address translation

Definitions

  • the present invention relates to a main storage managing method for computer systems and, more particularly, to a virtual storage managing method suited to use in a system in which a program requiring a high response speed and a program requiring a large storage capacity are executed simultaneously.
  • a virtual memory managing system is characterized in that the address of the memory device is not the real address (or physical address) imparted to the physical main storage but is identified by a virtual address attached to a virtual storage region which does not actually exist and is constituted solely of a logical construction. It is, therefore, essential to employ hardware for obtaining the correspondence between the virtual address and the real address as well as a software program for this purpose. The hardware and the software for obtaining this correspondence in combination are usually called in address translation mechanism.
  • the aforementioned DEMAND PAGING is one of the examples of a measure for realizing such an address translation mechanism.
  • the DEMAND PAGING METHOD therefore, will be outlined hereinbelow.
  • the virtual storage region is beforehand sectioned into a plurality of small regions of a fixed length referred to as a "Page".
  • the main memory is beforehand divided into small regions of the same length as the page and referred to as Page Frames (abbreviation PF, hereinunder).
  • the pages and PF are provided with Nos. starting from 0 (zero), respectively.
  • the program is so arranged on one or a plurality of consecutive pages in such a manner as not to lap the pages carrying other programs.
  • the following procedure is taken in the DEMAND PAGE METHOD for executing the program.
  • a reference is made to a virtual address in the page which is not correlated to the PF (this will be referred to as "page fault)
  • an operation is made to obtain correspondence between the page and the PF.
  • the aforementioned address translation mechanism operates to establish correpondence between the page and the PF which has not been correlated to the page (such PF will be referred to as "empty" PF).
  • empty PF When there is no empty PF, the correspondence between another page and PF is dismissed to make an empty PF and this new empty PF is made to correspond to the aforementioned page.
  • the size of the program is not limited by the capacity of the main storage. Namely, there is no limit in the size of the virtual storage region theoretically.
  • multiprogramming when a plurality of programs are executed simultaneously (referred to as "multiprogramming"), the total number of pages employed can be made much smaller as compared with the total number of pages of the program group.
  • the number of programs which can be contained by the main storage device is much greater than the number which is given by the total page number. This conveniently permits a degree of the multiplication of programs.
  • the third one is not so serious because the address translation mechanism can be nowadays produced at a sufficiently low cost thanks to the current development in the technology concerning hardware devices.
  • the first and second problems are fundamental ones and are serious particularly when the real time processing such as in plant control systems is conducted.
  • an object of the invention is to provide a virtual memory managing method which permits execution of a program requiring a high response speed, while maintaining the advantage of the DEMAND PAGE METHOD in which a large capacity program is conducted without being limited by the main memory capacity.
  • the virtual storage region which is the object of the address translation is divided into at least two parts: namely, an address translation fixed area (referred to as a "FX area”, hereinunder) in which the correspondence between the pages and PFs are made at the time of system generation, and an address translation variable region (Floating Area: referred to as a "FL" area, hereinunder) in which the correspondence between pages and PFs is established at the time of start of the program execution and is dismissed when the execution is over.
  • the programs requiring high response speed are disposed in the FX area, while programs of large capacities requiring no high response speed are disposed in the FL area.
  • the reason why also the FX area is used as the object of the address translation is as follows.
  • the address translation is fixed during the execution of the program.
  • the changed program is disposed in another virtual storage region.
  • the correspondence between the address and PFs is such that the pages carrying unchanged contents are alloted to the same PFs as previous ones, while the pages carrying changed or modified contents are alloted to the empty FSs.
  • FIG. 1 is an address translation chart for explaining a virtual storage managing system in accordance with the invention
  • FIG. 2 shows an example of a computer system to which the present invention is applied
  • FIG. 3 illustrates examples of an OS table and a program employed by the virtual storage managing system of the invention
  • FIG. 4 is an illustration of an example of the control table feed in accordance with the invention.
  • FIGS. 5A and 5B are flow charts of an area registration and deletion program
  • FIGS. 6A and 6B are flow charts of task forming and dismissing programs
  • FIGS. 7A and 7B are flow charts of task starting and finishing programs
  • FIG. 8 is a flow chart of a task dispatcher 4.
  • FIGS. 9A and 9B are flow charts of MS acquire and release programs.
  • FIG. 1 shows the arrangement of a virtual storage (referred to as "VS”, hereinunder) 100, main storage (referred to as “MS”, hereinunder) 300 and a secondary storage (referred to as “SS”, hereinunder) 400 in accordance with the invention, as well as correspondence of parts including address translation mechanism 200.
  • OS Operating System
  • FU Full Use
  • the V ⁇ R area which is an address translation area, is composed of an FX area in which the correspondence between the page and PF is obtained at the time of system generation and an FL area in which correspondence between the page and the PF is obtained at the time of start of program execution and is dismissed at the time of completion of the program execution.
  • the FX area is composed of a task RTASK (Resident Task) 150 and a task NRTASK (Non-Resident Task) 160, which are processing programs capable of performing a group of functions in a parallel manner, a global area (GLB) 130 which is a table usable commonly by the tasks, and a sub-routine (RSUB) 140 used commonly by the tasks.
  • RTASK Real Task
  • NRTASK Non-Resident Task
  • RSUB sub-routine
  • the area of the RTASK 150 is the area in which the pages are shared by the tasks without any overlap, and these tasks are loaded from the SS400 to the MS300 by the IPL410 at the time of start-up.
  • the NRTASK 160 is an area in which a plurality of tasks which are not conducted simultaneously or which need not be conducted simultaneously are made to correspond to the same VS address, thereby to make more efficient use of the VS.
  • the priority for the usage of this area is judged in accordance with the priority levels of these tasks. The judgement is made not at the time of start-up but at the time of execution of the task, and the tasks of inferior priority order are made to shelter into the shelter area 420 of the SS400.
  • the FL area also is composed of NRtask 170.
  • the MS300 is provided in a 1:1 relation to the VSs.
  • the FL area on the MS300 is smaller than the FL area on the VS.
  • Real addresses are alloted also for the MS starting with 0 (zero) from the left end in the ascending order for each byte.
  • the V ⁇ R region is divided by the units of the pages 180 and PFs 310. Numbers are alloted also for the pages 180 and PFs 310 in the ascending order starting with 0 (zero) from the left ends of VS and MS.
  • the SS400 is composed of an Initial Program Load (IPL) 410 which is a program adapted to store all regions excepting the FU120 of the VS100 program and to work out the informations of various regions on the MS from the state in which the content of the MS300 is unstable as in the case of turning ON and OFF of the power supply, a region for holding a file of a large capacity which cannot be stored in VS100 and a region for holding the SAVE AREA 420 for NRTASK160.
  • Address numbers are alloted also to the SS400 starting with 0 (zero) from the left end in ascending order for each 512 byte unit. This address will be referred to as a Logical Sector Address (abridged as "LSA”, hereinunder).
  • the VS100 carries three kinds of tasks having different managing methods, namely, the RTASK150, NRTASK160 and NRTASK170.
  • the RTASK150 of the FX area is used for a high-speed processing task of the most severe real time condition, while the NRTASK170 of the FL area is used for the tasks having sizes exceeding those of RTASK150 and NRTASK160.
  • the NRTASK160 of the FX area is used for the tasks of real time characteristics and capacity intermediate between those of the aforementioned two kinds of tasks.
  • FIG. 2 shows the whole hardware arrangement of a computer system to which the embodiment of the invention pertains.
  • the computer system is composed of the aforementioned MS300, a main storage control unit (referred to as "MCU") 500 for controlling the MS, a job processor (referred to as “JOBP”, hereinunder) having a function to read and execute the mechanical instruction on the MS, the aforementioned SS400, a file control electronic device (referred to as “FCE”, hereinunder) 890 for controlling the SS400, a file control processor (referred to as “FCP”, hereinunder) 800 for making a data transfer between the SS400 and MCU500 in accordance with the instructions given by the JOBP700 through the FCE890, a system BUS 610 which interconnects the MCU500,JOBP700 and the FCP800 and provides the data transfer means, and a system BUS controller (referred to as "SBC”, hereinunder) 600.
  • MCU main storage control unit
  • JOBP job processor
  • FCE file control electronic device
  • FCP file control processor
  • SBC system BUS controller
  • the JOBP700 is composed of a storage address register (referred to as "MAR”, hereinunder) 710 for instructing the MCU500 of the VS address as the access object, data register (referred to as “DR”, a hereinunder) 730 for storing the data for reading and writing, a group of flip-flops (referred to as “FF”, hereinunder) 720 for controlling the MCU500 and FCP800, an arithmetic operation unit (referred to as "ALU”, hereinunder) 740 having functions for logical operation, arithmetic operation and shift operation, a program counter (referred to as "PC”, hereinunder) 780 for showing the VS address storing the mechanical instruction to be executed next, a group of general registers (referred to as "GR”, hereinunder) 790 for use in the operation, a writable control storage (referred to as "WCS”, hereinunder) 750 for reading the mechanical instruction and storing what kind of processing should be made, a microprogram counter (referred to as "MPC”, here
  • the FCP800 has a construction similar to that of JOBP700 and corresponds to each of ALU740, WSC750, MPC760, ALU840, WCS850 and MPC860 but has no PC780 because it has no function for executing the mechanical instructions on the MS300. However, it has a transfer data buffer 870 for absorbing the difference in the data transfer speed between SS400 and MCU500, and an input/output control circuit 880 for controlling the transfer of data between itself and the FCE890 mentioned before.
  • VEQR area border register
  • PTIOR page table index origin register
  • the JOBP700 sets the VS address to be accessed in the MAR710 and sets information representing that the reference is made in the FF720.
  • the MCU500 takes up the VS address in the VAR530 and then starts up the ATU580.
  • the ATU580 determines the real address by making reference to the VEQR560, PTIOR570 and TLB510, and sets the same in the RAR520 thereby to start the MS300.
  • the ATU580 then picks up the data of address appointed by the RAR520 andsets the same in the input/output buffer register 550 and, thereafter, informs the FF540 of the fact that the data has been taken out.
  • the JOBP700 takes the data out of the input/output buffer register 550 into DR730 thereby to finish a series of operations for making reference to the MS.
  • FIGS. 3 and 4 an explanation will be made with specific reference to FIGS. 3 and 4 as to an address translation performed by the ATU580 using the hardware shown in FIG. 2 as the base.
  • FIG. 3 shows the construction of OS area 110 on the VS100.
  • the abbreviations used therein and their formal names are listed below.
  • the OS area is composed of a control table and a control program.
  • the control table has a multi-stage type list construction including an OS common block (OSCB) 10 disposed at the uppermost stage, under which are disposed page table index (PTIX) 20, page frame table (PFQ) 40, ready cupointerblock (RQPB) 50, task control block (TCB) 60, and an area control table (PCT) 70.
  • OSCB OS common block
  • PFQ page frame table
  • RQPB ready cupointerblock
  • TCB task control block
  • PCT area control table
  • a plurality of page tables (PT) 30 are disposed under the PTIX20, and, under the PFQ40, are disposed page frame table (PFT) 80 of list construction and TCB90.
  • TCB90 of list construction is disposed under RQPB50.
  • PTIOR Page Table Index Origin Register
  • TPPFQ Top Address of PFQ
  • TPRQPB Topic Address of RQPB
  • TPTCB Top Address of TCB
  • TPACT Top Address of ACT
  • the OSCB further includes the fields such as FXFLBV (FX,FL Boundary Address of VS) 17 representing the top VS address on the VS100, FXFLBM (FX,FL Boundary Address of MS) 18 which represents the top address of the FL area on the MS300, and MAXMS (Max Size of MS) 19 representing the final real address of the MS.
  • FXFLBV FX,FL Boundary Address of VS
  • FXFLBM FX,FL Boundary Address of MS
  • MAXMS Maximum Size of MS
  • the control program includes an area registration/deletion program 1 having a function for managing areas on the VS100, a task formation/dismissal program 2 for executing as tasks the RTASK150, NRTASK160,170 out of these areas, a task start/finish program having a function for starting and finishing the tasks formed by the program 2, a task dispatcher (Referred to as "DISP", hereinunder) having a function to select, out of a plurality of tasks to which the start demands are given, the one which is executable and having the highest level of priority and to shift the control to the selected task, and an MS acquire/release program 5 which is called by the DISP4 and, when the task is the NRTASK170, acquires and dismisses the correspondence between the pages 180 and the PFs 310.
  • DISP task dispatcher
  • FIG. 4 shows the construction of the TLB510 in the MCU500, as well as the constructions of the control tables PTIOR12, PTIX20, PT30, PFQ40, PQPB50, TCB60, ACT70 and the PFT80.
  • the abbreviations used therein and their formal names are listed below.
  • the VS address has a width of 32 bits (4 bytes) and the page size is 2048 bytes.
  • the TLB510 has 2 (two) sets of tables of 1024 entries. In each table, 1 entry corresponds to 1 page, and the access is made by detecting the entry No. of the TLB by the virtual page No. (referred to as "VPN", hereinunder) of 11th to 20th bits of the VS address.
  • VPN virtual page No.
  • This is constituted by the page frame No. PFN511, access protection information PRT512 for this page, VA513 corresponding to the first to tenth bits of the VS address (No. as obtained when the VS is sectioned for 1MB), a bit V514 for judging the validity of the TLB and the bit R515 for controlling the rewriting of two sets of TLB.
  • the PTIOR12 and the PTIX20 have an identical construction, and one entry of PTIX 20 is prepared for 1MB of VS area. This is constituted by a validity indication flag V21, TPPT(I) 22 representing the address of leading VS of the PTIX20 or PT30 and LENGTH 23.
  • the one entry of PT30 is prepared for one page.
  • the PT30 is composed of a validity indication flag 31, page frame No. PFN32 and an access protection information PRT33.
  • the Page Frame Queue Table PFQ40 is composed of the Page Frame Table (PFT) 80, the total case number PFCT41 of PFT, number EPFCT42 of vacant page frame, VS addresses (EFPNT,EBPNT) 43,44 of leading and final cases of vacant PFT, VS addresses (OFPNT,OBPNT) 45,46 of leading and final cases of occupied PFT and waiting line managing area (PFWFP,PEWBP) 47,48 for holding the task in the waiting state when it is judged that the vacant PFT80 is insufficient in the aforementioned MS acquired program 5.
  • Numerals 47 and 48 denote the addresses of the leading and final cases of the waiting TCB. All of the tables of the list construction are managed by means of the VS addresses of the leading case and the final case as in the present table.
  • RQPB 50 is prepared for each preferential level of the task constituted by the waiting line headers 51,52 of the TCB60.
  • the aforementioned task start up program 3 connects the TCB60 of the task to be started to the trailing end of the line waiting for the execution under the control of the managing table which is the RQPB corresponding to the preferential level determined by the task to be started.
  • the task completion program 3 gets rid of the TCB60 from the line waiting for the execution, when the execution of the task is completed.
  • the TCB60 is composed of pointers RQFP61,RQBP62 prepared for each task and adapted for connection to the RQPB50, resource waiting pointers 63,64 connected in the case of the resource waiting condition for the resources such as PFWFP47,PFWBP48 of PFQ40, task No. 64, waiting flag E66-1, other control flag 66, preferential flag 67-1, access protection information (PRT) 67 and other control informations 68.
  • PRT access protection information
  • the ACT70 is prepared for each module (each RSUB,GLB, task and so forth) in each area on VS100, and is composed of pointers ACTFP71,ACTBP72 for seeking the VS address of the next table, module identifying information IDENTFIER 73, leading VS address VA74 on which the module is disposed, leading SS address LSA75, module size (number of bytes) SIZE76 and other control informations.
  • the number of the PFT80 prepared corresponds to the number of the PF310 in the FX area on the aforementioned MS300.
  • Each PFT80 is composed of pointers FPNT81,BPNT82 for searching for the VS addresses of the next table, page No. VPN83 and page frame No. PFN84.
  • the entry of the TLB510 is searched out from the 11th to 20th bits of the VAR 530, and a check is made as to whether the first to tenth bits coincide with the VA513 of either one of two sets of the TLB510.
  • TLB hit a coincidence is confirmed (referred to as "TLB hit", hereinunder)
  • the PFN511 of the TLB 510 is set in the 9th to 20th bits in the RAR520, and the 31st bits of the VAR are set in the 21st to 31st bits in the RAR 520 thus completing the address translation.
  • the process proceeds to the next step.
  • the leading address of PTIX20 is determined from the TPPT(I) 22 of the aforementioned PTIOR570.
  • the LENGTH23 is compared with the first to seventh bits of the VAR 530 and, if the latter is smaller, this condition is judged to be the page fault mentioned before, because it means that there is no PTIX20. If the LENGTH23 is equal to or greater than the content of the first to seventh bits of the VAR530, the address of the case corresponding to the PTIX20 is determined from the first to 11th bits of the VAR530 and the V bit 21 is judged.
  • the state is judged to be page fault if the V bit is judged to be invalid. If valid, the leading address of the PT30 is determined by the TPPT(I) 22. Then, a comparison is made between the 12th to 16th bits of the VAR530 and the LENGTH23 of the PTIX20 to check for the length of PT30 as in the case explained before. Then, the address of the case corresponding to the PT30 is determined from the 12th to 20th bits of the VAR530 and a judgement is made as to whether the V bit 31 is valid. If the V bit 31 proved to be invalid, the state is judged to be a page fault. If valid, the PFN32 is set in the 9th to 20th bits of the RAR520 while the following 21st to 31st bits receive the 21st to 31st bits of the VAR530 thus completing the address translation.
  • FIGS. 5 to 13 in combination show the flow charts of the control programs 1 to 5 shown in FIG. 3.
  • FIGS. 5A and 5B show the flow chart of the area registration and deletion program 1.
  • the area registration program A performs the inputting of the No. for identifying the kind of the area, leading VS address, leading SS address, capacity and control information, and performs an operation (902,903) for setting them in the ACT70.
  • the area deletion program B performs the inputting (904) of the kind and identification No. of the area, as well as operation (905,906) for making the ACT70 empty.
  • FIGS. 6A and 6B show the flow charts of the task formation/dismissal program 2.
  • the task formation program A performs the inputting (910) of the task No., preference level and protection information, as well as setting (911 to 915) of the same in the TCB60.
  • the task deletion program B performs the inputting (916) of the task No. and operation (917,918) for making the TCB60 empty.
  • FIGS. 7A to 7B show flow charts of the task starting and finishing program 3.
  • the task starting program A makes the inputting (920) of the task No. and determination of preference level of the corresponding task (921,923).
  • the task starting program A further conducts an operation (923,924) for connecting the TCB60 to the waiting line of the preference level managed by the RQPB50.
  • the task finishing program B makes operations such as determination (925) of the preference level from the TCB60 of its own, isolation (926,927) of the same from the execution waiting line and shifting (928) of the control to the task dispatcher 4 before the control is changed to another task.
  • FIG. 8 shows the flow chart of the task dispatcher 4.
  • This program makes a search (930,931,933,934) for the TCB60 sequentially from the execution waiting line of the RQPB50 of the highest preference level and seeks out (932) the one E flag 66-1 which is OFF.
  • This program then makes judgement (935,936,937) to determine whether this task belongs to the FL area. If so, the program calls for the MS acquired program 5 (937,938) and, if the finish is made in the normal way, conducts an operation (941) for shifting the control to this task.
  • FIGS. 9A and 9B show the flow charts of the MS acquire release program 5.
  • the MS acquire program A makes operations such as inputting (950) of the task No., leading address of the VS area and the page No. and a judgement (951) as to whether there is sufficiently ample empty PFT80. If there is sufficient empty PTF, the program determines the PTF80 and makes the same occupied (952,953) and sets (955) the VPN83 while making operation (956,957) for working out corresponding PT30 for all pages (958,959) to permit a normal finishing. When the above-mentioned judgement by PFQ proved insufficient number of PFT80, the program A holds (961,962,963) in the waiting condition, and performs an operation (964) for reporting shortage of the frame number.
  • the MS release program B inputs (970) the leading address of the VS area and the page No. and makes an operation (971,972,974,975,976) for making the corresponding PFT80 vacant.
  • the program B then makes a judgement (977) as to whether there is any TCB waiting for empty page frame. If any, it makes an operation (978) for dismissing the tasks from the waiting condition.
  • the task finishing program (FIG. 7B), task deletion program (FIG. 6B) and the area deletion program (FIG. 5B) are executed in a timed relation in the mentioned order.
  • the task finishing program (FIG. 7B)
  • task deletion program (FIG. 6B)
  • the area deletion program (FIG. 5B)
  • the address fix area of the address translation area is the area in which the relationship or correspondence between the virtual address and the physical address is fixed.
  • the real time tasks requiring high response speed can be executed by making use of this address fix area.
  • the address variable area is an area which dynamically allots and operates the real space necessary in the start up of the task.
  • the program which does not necessitate real time operation, can be executed by making use of this area, without being restricted by the address, i.e. without being limited by the capacity of the main storage.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Memory System Of A Hierarchy Structure (AREA)
US06/448,274 1981-12-11 1982-12-09 Virtual storage management Expired - Lifetime US4563737A (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP56-200645 1981-12-11
JP56200645A JPS58102380A (ja) 1981-12-11 1981-12-11 仮想記憶管理方法

Publications (1)

Publication Number Publication Date
US4563737A true US4563737A (en) 1986-01-07

Family

ID=16427835

Family Applications (1)

Application Number Title Priority Date Filing Date
US06/448,274 Expired - Lifetime US4563737A (en) 1981-12-11 1982-12-09 Virtual storage management

Country Status (4)

Country Link
US (1) US4563737A (de)
EP (1) EP0081822B1 (de)
JP (1) JPS58102380A (de)
DE (1) DE3280414T2 (de)

Cited By (18)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4860192A (en) * 1985-02-22 1989-08-22 Intergraph Corporation Quadword boundary cache system
US4884197A (en) * 1985-02-22 1989-11-28 Intergraph Corporation Method and apparatus for addressing a cache memory
US4899275A (en) * 1985-02-22 1990-02-06 Intergraph Corporation Cache-MMU system
US4933835A (en) * 1985-02-22 1990-06-12 Intergraph Corporation Apparatus for maintaining consistency of a cache memory with a primary memory
US4937735A (en) * 1987-09-28 1990-06-26 Nec Corporation Memory access system utilizing address translation
US5091846A (en) * 1986-10-03 1992-02-25 Intergraph Corporation Cache providing caching/non-caching write-through and copyback modes for virtual addresses and including bus snooping to maintain coherency
US5113522A (en) * 1989-05-17 1992-05-12 International Business Machines Corporation Data processing system with system resource management for itself and for an associated alien processor
US5144692A (en) * 1989-05-17 1992-09-01 International Business Machines Corporation System for controlling access by first system to portion of main memory dedicated exclusively to second system to facilitate input/output processing via first system
US5155809A (en) * 1989-05-17 1992-10-13 International Business Machines Corp. Uncoupling a central processing unit from its associated hardware for interaction with data handling apparatus alien to the operating system controlling said unit and hardware
US5255384A (en) * 1985-02-22 1993-10-19 Intergraph Corporation Memory address translation system having modifiable and non-modifiable translation mechanisms
US5282274A (en) * 1990-05-24 1994-01-25 International Business Machines Corporation Translation of multiple virtual pages upon a TLB miss
US5283868A (en) * 1989-05-17 1994-02-01 International Business Machines Corp. Providing additional system characteristics to a data processing system through operations of an application program, transparently to the operating system
US5325517A (en) * 1989-05-17 1994-06-28 International Business Machines Corporation Fault tolerant data processing system
US5329629A (en) * 1989-07-03 1994-07-12 Tandem Computers Incorporated Apparatus and method for reading, writing, and refreshing memory with direct virtual or physical access
US5369749A (en) * 1989-05-17 1994-11-29 Ibm Corporation Method and apparatus for the direct transfer of information between application programs running on distinct processors without utilizing the services of one or both operating systems
US5369767A (en) * 1989-05-17 1994-11-29 International Business Machines Corp. Servicing interrupt requests in a data processing system without using the services of an operating system
US6026475A (en) * 1997-11-26 2000-02-15 Digital Equipment Corporation Method for dynamically remapping a virtual address to a physical address to maintain an even distribution of cache page addresses in a virtual address space
US6148387A (en) * 1997-10-09 2000-11-14 Phoenix Technologies, Ltd. System and method for securely utilizing basic input and output system (BIOS) services

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19717102A1 (de) * 1997-04-23 1998-10-29 Siemens Nixdorf Inf Syst Verfahren zur Übertragung eines Betriebssystems in Datenverarbeitungsanlagen
US7302546B2 (en) 2004-01-09 2007-11-27 International Business Machines Corporation Method, system, and article of manufacture for reserving memory
GB2460462A (en) * 2008-05-30 2009-12-02 Symbian Software Ltd Method for loading software components into RAM by modifying the software part to be loaded based on the memory location to be used.

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3825904A (en) * 1973-06-08 1974-07-23 Ibm Virtual memory system
US3902164A (en) * 1972-07-21 1975-08-26 Ibm Method and means for reducing the amount of address translation in a virtual memory data processing system
US3964028A (en) * 1973-08-02 1976-06-15 International Business Machines Corporation System and method for evaluating paging behavior
US4037215A (en) * 1976-04-30 1977-07-19 International Business Machines Corporation Key controlled address relocation translation system
US4084225A (en) * 1976-09-24 1978-04-11 Sperry Rand Corporation Virtual address translator
US4096568A (en) * 1976-09-24 1978-06-20 Sperry Rand Corporation Virtual address translator

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3914747A (en) * 1974-02-26 1975-10-21 Periphonics Corp Memory having non-fixed relationships between addresses and storage locations
JPS5821307B2 (ja) * 1976-09-14 1983-04-28 横河電機株式会社 アドレス変換機能を持つたデ−タ処理装置
JPS56119981A (en) * 1980-02-25 1981-09-19 Mitsubishi Electric Corp Address converting circuit

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US3902164A (en) * 1972-07-21 1975-08-26 Ibm Method and means for reducing the amount of address translation in a virtual memory data processing system
US3825904A (en) * 1973-06-08 1974-07-23 Ibm Virtual memory system
US3964028A (en) * 1973-08-02 1976-06-15 International Business Machines Corporation System and method for evaluating paging behavior
US4037215A (en) * 1976-04-30 1977-07-19 International Business Machines Corporation Key controlled address relocation translation system
US4084225A (en) * 1976-09-24 1978-04-11 Sperry Rand Corporation Virtual address translator
US4096568A (en) * 1976-09-24 1978-06-20 Sperry Rand Corporation Virtual address translator

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
S. E. Madnick & J. J. Donovan: Operating Systems, McGraw Hill Book Co., 1974, pp. 139 165. *
S. E. Madnick & J. J. Donovan: Operating Systems, McGraw-Hill Book Co., 1974, pp. 139-165.

Cited By (22)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5255384A (en) * 1985-02-22 1993-10-19 Intergraph Corporation Memory address translation system having modifiable and non-modifiable translation mechanisms
US4884197A (en) * 1985-02-22 1989-11-28 Intergraph Corporation Method and apparatus for addressing a cache memory
US4899275A (en) * 1985-02-22 1990-02-06 Intergraph Corporation Cache-MMU system
US4933835A (en) * 1985-02-22 1990-06-12 Intergraph Corporation Apparatus for maintaining consistency of a cache memory with a primary memory
US4860192A (en) * 1985-02-22 1989-08-22 Intergraph Corporation Quadword boundary cache system
US5091846A (en) * 1986-10-03 1992-02-25 Intergraph Corporation Cache providing caching/non-caching write-through and copyback modes for virtual addresses and including bus snooping to maintain coherency
US4937735A (en) * 1987-09-28 1990-06-26 Nec Corporation Memory access system utilizing address translation
US5369749A (en) * 1989-05-17 1994-11-29 Ibm Corporation Method and apparatus for the direct transfer of information between application programs running on distinct processors without utilizing the services of one or both operating systems
US5363497A (en) * 1989-05-17 1994-11-08 Ibm Corporation System for removing section of memory from first system and allocating to second system in a manner indiscernable to both operating systems
US5144692A (en) * 1989-05-17 1992-09-01 International Business Machines Corporation System for controlling access by first system to portion of main memory dedicated exclusively to second system to facilitate input/output processing via first system
US5701502A (en) * 1989-05-17 1997-12-23 International Business Machines Corporation Isolating a central processing unit from the operating system controlling said unit and its associated hardware for interaction of the unit with data handling apparatus alien to the operating system
US5283868A (en) * 1989-05-17 1994-02-01 International Business Machines Corp. Providing additional system characteristics to a data processing system through operations of an application program, transparently to the operating system
US5325517A (en) * 1989-05-17 1994-06-28 International Business Machines Corporation Fault tolerant data processing system
US5388215A (en) * 1989-05-17 1995-02-07 Ibm Corporation Uncoupling a central processing unit from its associated hardware for interaction with data handling apparatus alien to the operating system controlling said unit and hardware
US5155809A (en) * 1989-05-17 1992-10-13 International Business Machines Corp. Uncoupling a central processing unit from its associated hardware for interaction with data handling apparatus alien to the operating system controlling said unit and hardware
US5113522A (en) * 1989-05-17 1992-05-12 International Business Machines Corporation Data processing system with system resource management for itself and for an associated alien processor
US5369767A (en) * 1989-05-17 1994-11-29 International Business Machines Corp. Servicing interrupt requests in a data processing system without using the services of an operating system
US5329629A (en) * 1989-07-03 1994-07-12 Tandem Computers Incorporated Apparatus and method for reading, writing, and refreshing memory with direct virtual or physical access
US5282274A (en) * 1990-05-24 1994-01-25 International Business Machines Corporation Translation of multiple virtual pages upon a TLB miss
US6148387A (en) * 1997-10-09 2000-11-14 Phoenix Technologies, Ltd. System and method for securely utilizing basic input and output system (BIOS) services
US6892304B1 (en) 1997-10-09 2005-05-10 Phoenix Technologies Ltd. System and method for securely utilizing basic input and output system (BIOS) services
US6026475A (en) * 1997-11-26 2000-02-15 Digital Equipment Corporation Method for dynamically remapping a virtual address to a physical address to maintain an even distribution of cache page addresses in a virtual address space

Also Published As

Publication number Publication date
EP0081822B1 (de) 1992-09-09
JPS58102380A (ja) 1983-06-17
EP0081822A2 (de) 1983-06-22
DE3280414T2 (de) 1993-03-25
JPH0313615B2 (de) 1991-02-22
DE3280414D1 (de) 1992-10-15
EP0081822A3 (en) 1986-07-16

Similar Documents

Publication Publication Date Title
US4563737A (en) Virtual storage management
US4577274A (en) Demand paging scheme for a multi-ATB shared memory processing system
EP0856797B1 (de) Cachespeichersystem für gleichzeitig laufende Prozesse
US5539899A (en) System and method for handling a segmented program in a memory for a multitasking data processing system utilizing paged virtual storage
KR920005853B1 (ko) 페이지 분할형 가상 메모리 데이타 처리 시스템에서의 데이타 전송 방법
US5940872A (en) Software and hardware-managed translation lookaside buffer
US5119290A (en) Alias address support
EP0431467B1 (de) Multiprozessorsystem mit verteilten gemeinsamen Betriebsmitteln und mit Vervielfältigung von globalen Daten
US3938096A (en) Apparatus for developing an address of a segment within main memory and an absolute address of an operand within the segment
US6256715B1 (en) System and method of performing gateway access
EP0447736A1 (de) Mehrrechnersystem mit verteilten gemeinsamen Betriebsmitteln und dynamischer und selektiver Vervielfältigung globaler Daten und Verfahren dafür
US7493464B2 (en) Sparse matrix
JPS61141055A (ja) 情報処理装置のアドレス変換方式
US5293599A (en) Process for partially swapping real storage areas used by a program between a real storage and an auxiliary storage
JP2768503B2 (ja) 仮想記憶アドレス空間アクセス制御方式
US5619691A (en) File sharing method and system for multiprocessor system
EP1103898A2 (de) Mikroprozessor und Speicher
US4984150A (en) Virtual memory control management system
US5450587A (en) Expanded memory addressing scheme
JPH0479019B2 (de)
JP3456727B2 (ja) データ処理装置
Wilkes The use of a writable control memory in a multiprogramming environment
Welch An investigation of descriptor oriented architecture
JPH09282233A (ja) 仮想メモリ管理装置
JPH0887443A (ja) セルプール領域管理処理方法

Legal Events

Date Code Title Description
AS Assignment

Owner name: HITACHI, LTD., 5-1, MARUNOUCHI 1-CHOME, CHIYODA-KU

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST.;ASSIGNORS:NAKAMURA, TOMOAKI;NAKANE, KEIICHI;NAKANISHI, HIROAKI;AND OTHERS;REEL/FRAME:004430/0563

Effective date: 19821129

STCF Information on status: patent grant

Free format text: PATENTED CASE

FPAY Fee payment

Year of fee payment: 4

FEPP Fee payment procedure

Free format text: PAYOR NUMBER ASSIGNED (ORIGINAL EVENT CODE: ASPN); ENTITY STATUS OF PATENT OWNER: LARGE ENTITY

FPAY Fee payment

Year of fee payment: 8

FPAY Fee payment

Year of fee payment: 12